Multivariate models as predictors of pathological stage using gleason score, clinical stage, and serum prostate-specific antigen

T. J. Polascik, J. D. Pearson, A. W. Partin

Research output: Contribution to journalReview articlepeer-review

14 Scopus citations


The patient presented is a 58-year-old man with newly diagnosed prostate cancer who likely has at least a 10-to 15-year life expectancy. In staging this man's extent of disease, several preoperative clinical variables are provided to assess whether the patient has local disease before offering definitive surgical therapy. It has been demonstrated that the combination of several of these variables in a multivariate analysis has greater predictive power than any of these variables do alone. Multivariate analysis using clinical stage, prostate-specific antigen (PSA), and Gleason score will provide both the physician and patient with 95% confidence intervals for determining the probability of having organ-confined disease, extracapsular penetration, seminal vesicle involvement, and lymphatic metastases. Several of the clinical variables given indicate that this patient has advanced disease, such as a PSA of 12 ng/mL, a stage T2b lesion, Gleason sum 7 disease in 2 of 3 cores from the right side associated with perineural invasion, and an additional Gleason sum 7 biopsy from the contralateral apex. For a man with these preoperative variables, there is a 13% chance of organ-confined disease, an 18% probability of seminal vesicle invasion, and a 17% chance of positive pelvic lymph nodes based on a nomogram constructed from multivariate analysis. Using this information, this man should be counseled that he has a high probability (87%) of extracapsular disease and a significant risk (15% to 20%) of having either seminal vesicle or lymph node involvement. Recognizing the risks and benefits of various forms of definitive therapy for prostate cancer, the patient has additional information to make an informed decision.

Original languageEnglish (US)
Pages (from-to)160-171
Number of pages12
JournalSeminars in Urologic Oncology
Issue number3
StatePublished - 1998


  • Multivariate analysis
  • Nomogram
  • Prostate cancer
  • Prostate-specific antigen

ASJC Scopus subject areas

  • Urology


Dive into the research topics of 'Multivariate models as predictors of pathological stage using gleason score, clinical stage, and serum prostate-specific antigen'. Together they form a unique fingerprint.

Cite this